Premature rupture of the fetal membrane (amniotic sac) occurs in about 10% of pregnant women and when not treated promptly, it is the cause of about 10% of all perinatal deaths. The term PROM (premature rupture of the fetal membranes) relates to the spontaneous rupture of the membranes 24 or more hours before the onset of labor either at term or preterm. PPROM refers to preterm premature rupture of membranes. Approximately about 30-50% of such premature ruptures occur before the 37th week of pregnancy. In such cases, definitive diagnosis of the rupture is extremely important since PROM is associated with a significant increase in the risk of an intrauterine infection and disturbance of development of the fetal lung system. Intrauterine penetration of such infections increases both maternal and perinatal morbidity and mortality by about ten percent. Immediate diagnosis of a rupture at 38 to 40 weeks of pregnancy is crucial, since once PROM is detected delivery should be induced as soon as possible. The rupture diagnosis is also important before 37 weeks of pregnancy because it enables prevention of intra-amnion infection and the stimulation of fetal lung development.
There is no “gold standard” available for the diagnosis of membrane rupture. PROM is a dynamic entity, so the interval between membrane rupture and implementation of the diagnostic modality, the presence of “high” leaks, intermittent leakage, variations in the incidence of PROM relative to populations, and consideration of material that has the capability of interfering with test results are factors that when not addressed result in inaccurate reporting. These inaccuracies may lead to errors in interpreting studies which aim to reveal the best tool for the identification of PROM.
The diagnosis of PROM has traditionally relied on the patient's report of fluid discharge from the vagina. Physical examination has the capability to diagnose unequivocally; however, there are times when the findings at examination are internally inconsistent or equivocal. This situation mandates the need for confirmatory diagnostic tests (Lockwood C. J. et al., Am. J. Obstet. Gynecol., 1994, v. 171, No 1, pp. 146-150). Several methods, all of them insufficient, are presently used to detect amniotic fluid in the vagina, such as the fern test (M. L. Friedman and T. W. McElin, “Diagnosis of Ruptured Fetal Membranes”, American Journal of Obstetrics and Gynecology 1969, Vol. 100, pp. 544-550). This method is based on the detection of the amniotic fluid by the observation of so-called arborization when the amniotic fluid dries on a slide. This method, however, is not sufficiently accurate since it is based on the highly volatile properties of amniotic fluid in the vagina. It may produce false results in as many as 30 percent of the cases.
It has been also proposed to detect the rupture of the fetal membrane by employing several dyes: nile blue, acridine orange, bromthymol blue, nitrazine, etc. (M. L. Friedman and T. W. McElin, supra). This approach is inconvenient and has disadvantages related to the volatility of the chemical properties of amniotic fluid in the vagina and some possible admixtures to it. For instance, a vaginal infection can influence the results of the above tests. An early study of currently prevalent Nitrazine and Ferning tests indicated that these tests had high inaccuracy rates, which increased progressively when more than one hour has elapsed since membrane rupture, and became inconclusive after 24 hours. The study concludes that in cases of prolonged PROM these tests provide no better diagnostic information than that obtained by simple clinical evaluation (Gorodeski I. G, Haimovitz L., Bahari C. M., Journal Perinat. Med, 1982, v. 10, No 6, pp. 286-292). More recent data (Trovo S. et al., Minerva Ginecol. 1998, v. 50, No 12, pp. 519-512) on the tests are:
Nitrazine test shows sensitivity 70%, specificity 97%, accuracy 90%;
Ferning test shows sensitivity 70%, accuracy 93%.
It has been proposed recently to detect the rupture of fetal membranes based on an immunochemical analysis of the proteins in the amniotic fluid. Docked immunochemical analysis utilizes the following proteins of the amniotic fluid to detect a membrane rupture: alpha-fetoprotein, prolactin, fibronectin, and insulin-like growth-factor binding protein 1, see B. L. Rochelson et al, “Rapid Assay—Possible Application in the Diagnosis of Premature Rupture of the Membranes”, in Obstetrics and Gynecology, 1983, v. 62, pp. 414-418; P. R. Koninckx et al., “Prolactin Concentration in Vaginal Fluid: a New Method for Diagnosing Ruptured Membranes”, British J. Obstetr. Gynecol., 1981, v. 88, pp. 607-610; P. Hellemans, et al., “Preliminary Results with the Use of the ROM Check Immunoassay in the Early Detection of Rupture of the Amniotic Membranes”, Eur. J. Obstet. Gynecol. Reprod. Biol., 1992, v. 43(3), pp. 173-179; Rutanen, E. M., et al., “Measurement of Insulin-like Growth-Factor binding Protein-1 in Cervical/Vaginal Secretions: Comparison with the ROM Check Membrane Immunoassay in the Diagnosis of Ruptured Fetal Membranes”, Clin. Chim. Acta., 1993, v. 214, pp. 73-81. Rutanen, E. M., et al. developed later a chromatographic test using the upside-down-positioned chromatographic membrane (FI-84863; U.S. Pat. No. 5,554,504).
The methods which are based on the detection of alpha-fetoprotein (AFP) and prolactin (PRL) are unreliable since the blood/amniotic fluid ratio of AFP and PRL proteins is prone to significant variations. AFP and PRL are present in amniotic fluid in high concentrations during the second trimester of pregnancy only. The amniotic/serum protein concentration ratio for both proteins is only about 3 to 4 at term.
Another method based on the detection of fetal fibronectin in the vaginal secretions has also been found unsatisfactory. For instance, the presence of fetal fibronectin can take place even in the absence of the fetal membrane rupture (P. Hellemans, at al., “Preliminary Results with the Use of the ROM Check Immunoassay in the Early Detection of Rupture of the Amniotic Membranes”, Eur. J. Obstetr. Gynecol. Reprod. Biol. 1992, v. 43(3), pp. 173-179; C. Lockwood, et al., “Fetal Fibronectin in Cervical and Vaginal Secretions as a Predictor of Preterm Delivery”, New England Journal of Medicine, 1991, v. 325, pp. 669-674), thereby producing false-positive results.
All of these methods of detecting fetal membrane rupture, based on detection of alpha-fetoprotein, prolactin, and fibronectin, are inaccurate due to variable factors in control of the concentration of these proteins in amniotic fluid and of the relative concentration of these proteins in the amniotic fluid to that in blood serum.
As for the IGFBP-1 test update, there are contradictory data concerning its specificity and accuracy. A rapid strip test (PROM test by OY Medix Biochemica, Finland, also named Amni-check, MAST Diagnostica, Germany), has been developed for detecting the presence of IGFBP-1 in the vaginal secretions (Rutanen E M, Karkkainen T H, Lehtovirta J., Uotila J T, Hinkula M K, Hartikainen A L. “Evaluation of a rapid strip test for insulin-like growth factor binding protein-1 in the diagnosis of ruptured fetal membranes”, Clin Chim Acta 1996 September 30; v. 253(1-2), pp. 91-101). E. Rutanen reported that the detection limit of the test was set so that IGFBP-1 concentrations below 400 ng/ml in cervical secretion (below the 95th percentile of serum IGFBP-1 levels in pregnant women) should remain negative. However, in cases with bleeding, the test result should be interpreted with caution as blood straight from the placental bed may contain higher amounts of IGFBP-1 than blood from the cervical blood vessels.
All samples (n=55) in women with clinically confirmed PROM showed a positive result and 71 of 75 samples taken from asymptomatic women were negative according to the test. Among this set of samples, the test had sensitivity of 100% and specificity of 94.7%. This fact can be explained by insufficient specificity (cross reactivity) of the monoclonal antibody used at the first step of testing.
Among the 181 patients evaluated for suspected, but upon initial examination equivocal PROM, the test was positive in 64 cases and negative in 117 cases. Fifty of 64 positive patients (78.1%) delivered before 37 weeks of gestation, 42 (65.6%) within 2 weeks after sampling. Five of 117 patients with a negative test result had elective cesarean section for reasons unrelated to PROM. Among the other 112 patients, 102 (91.1%) delivered at term and 10 (8.9%) delivered before 37 weeks, seven of those (6.3%) within two weeks after sampling (E. Rutanen et al. 1996). Unfortunately, there is no data regarding sensitivity and specificity of the PROM test in women with unequivocal diagnosis of PROM.
In a study by W. Woltmann, Amni-check was used to detect IGFBP-1 in 150 amniotic fluid specimens and 50 vaginal secretion samples from women with clinically unconfirmed PROM. The test had a sensitivity of 97% and a specificity of 100% (Woltmann W. et al., Z. Gebursh. Neonatal, 1995, v. 199, pp. 243-244).
V. Ragosh evaluated diagnostic accuracy of the Amni-check test in the 75 vaginal secretion samples. The test showed a sensitivity of 100% and a specificity of 83%. Investigators reported that the false positive rate was strongly dependent on the labor activity. In women with uterine contractions, the test had a specificity of 59% (Ragosch, V. et al., Geburtsh. U. Frauenheililk., 1996, Vol. 56, pp. 291-296).
In a study by E. Darj and S. Lyrenas (Acta Obstet. Gynecol. Scand., 1998, v. 77, pp. 295-297), PROM-test had a sensitivity of 95.7% and a specificity of 93.1% among the patients with clinically confirmed diagnosis (women with obvious rupture of membranes or women with intact membranes). However, the sensitivity and specificity of PROM-test were only 70.8% and 88.2% respectively in the patients with suspected PROM. This discrepancy could be explained by the cut-off limit of the test (400 ng/ml), which makes it impossible to detect a small amount of amniotic fluid in vaginal secretions of patients with equivocal diagnosis (for instance, in the case of a small rupture).
Thus a significant background level of vaginal IGFBP-1 in women with intact membranes and a high cutoff threshold of the test may harm its sensitivity and specificity and thus impact the test's accuracy in patients with equivocal diagnosis. The admixtures of blood serum and/or inflammatory exudate could also impact the accuracy of test (see data of E. Darj et S. Lyrenas, above). The author of this test did not study the issue.
In attempting to avoid some of the above-mentioned drawbacks, two monoclonal antibodies were used against two binding sites for insulin-like growth factors to detect the unbound fraction of placental α1-microglobulin (U.S. Pat. Nos. 5,968,758; 5,597,700; 5,891,722; 5,877,029).
In these patents the identity of the two proteins, unbound PAMG-1 and IGFBP-1, was baselessly assumed. As a matter of fact, such assumption could be based only on the comparison of the primary structure and genes of these proteins.
In the above-mentioned patents it was not possible to set up the threshold of sensitivity of such test so as to achieve the highest degree of accuracy possible (99% or above). The common problem for such tests is background level and variability of background concentration of the detected substance. For instance, the background level of another protein, IGFBP-1, in the vaginal secretion of pregnant women, varies in a broad range from 0.5 to 90 ng/ml (see Rutanen's studies). The second important point is the possibility of admixtures of inflammation exudates or blood serum containing detected substance in vaginal secretion. This can cause false positive results.
Protein PAMG-1 was first described by D. Petrunin (Petrunin D. et al, Akusherstvo i Ginekologia, 1977, No. 1, p. 64, in Russian; see also PMID:65924 (PubMed-indexed for MEDLINE: “Immunochemical identification of organ specific human placental alpha-globulin and its concentration in amniotic fluid”, Akusherstvo i Ginekologia (Moscow) 1977 January, Vol. 1, p. 64)). Antibodies were obtained against the purified and isolated protein, and immunochemical methods permitted measuring the contents of the protein in amniotic fluid (including amniotic fluid taken from the vagina) at different stages of pregnancy. The concentration of the protein in blood and different organs of the fetus and adult was also measured.
This research group continued to publish new results on the protein during subsequent years, until 1990 (Petrunin, D. et al, “Comparative Study of Four Placental Protein During Gestation”, Akusherstvo i Ginekologia, 1988, No. 1, pp. 50-52; Zaraisky, E. et al, Voprosy Med. Khemii, 1989, No 5, pp. 131-132; Tatarinov, Y. et al, Uspekhi Sovr. Biologii 1990, Vol. 109, pp. 369-373; Boltovskaya, M. et al, Bulletin of Experimental Biology and Medicine, 1991, No. 7, pp. 397-400; Nasimova, S. V. et al, Bulletin of Experimental Biology and Medicine, 1993 September; Vol. 116, No. 9, pp. 302-304 (all these papers are in Russian with English abstracts). D. Petrunin obtained the Invention Certificate on the method of isolation of PAMG-1 (# SU-1614184 A1, Priority year 1988).
In 1988-89 a few papers were published detailing the partial and full sequence of similar proteins, the Insulin-like Growth Factor Binding Proteins (IGFBP), obtained from the amniotic fluid, from placenta and from human hepatoma (Bell S et al, 1988; Luthman H. et al, 1989; Julkunen et al, 1988; Lee, Y. et al, 1988). The gene was localized in the piece 7p14-7p12 of the 7th human chromosome. Before 1991, researchers used different names for this protein: α1-PEG, PP-12, IGFBP, BP-25, etc.
In 1980-82, Bohn isolated a protein from the placenta and called it PP-12. In his paper, he compared PP-12 to the PAMG-1 protein, discovered earlier, and discussed the similarities and differences between them.
In contrast to the other research publications was a paper by Bell et al (1988), who found polymorphism in the N-end peptide of the α1-PEG protein, namely in the in the 11th and 12th positions, and came to the conclusion that there were, in actuality, two different proteins rather than one.
S. Bell once again references his own paper regarding the two different proteins α1-PEG in amniotic fluid. This paper accepts the decision of the Nomenclature Committee of 1990 (Report on the Nomenclature of the IGF Binding Proteins, Journ. Clin. Endocr. And Metabol. 1990, 70, #3, p. 817), which decided that proteins AFBP, PP-12, α1-PEG, GH-Protein, Binding Proteins 28, 26, 25, JB-1 are identical and gave them all a general name hIGFBP-1.
A so-called free PAMG-1 was used to detect the fetal membrane rupture. However, as mentioned above, a test with high accuracy (>99%) was not created. This goal was achieved later with our new method and device, described in this Application. The present invention employs a method of selection of a pair of monoclonal antibodies to provide sensitivity sufficient to detect a very low concentration of PAMG-1 in the vaginal secretion, and also involves selection of some other anti-PAMG-1 antibodies, which in combination with the two antibodies mentioned above, allowed to precisely set up a predefined threshold of sensitivity for the strip device. This, in turn, made it possible to minimize the frequency of false positive results of the test.
The present invention started from the pioneer study D. Petrunin, who separated and described placental-alpha-microglobulin and carried out a thorough measurement of its concentration in the amniotic fluid, blood and some tissues using immunochemical methods. This publication is the public domain that should be taken into account by any researcher. The method of separation of PAMG-1 has been protected by an official author's certificate (# SU-1614184 A1, Priority year 1988), an equivalent of a patent in the former USSR.